Re: [PATCH 07/11] PCI: rcar-gen4: Recover the Root Port on link down

[Koichiro Den]

On Tue, Sep 22, 2026 at 11:44:02PM +0200, Marek Vasut wrote:
> On 9/18/26 5:20 AM, Koichiro Den wrote:
> > On R-Car, intreq_pcim_sub carries both the integrated MSI receiver and
> > the controller's reset requests (smlh_req_rst_not, link_req_rst_not), so
> > the generic DesignWare chained handler reads the MSI status from DBI as
> > soon as the link goes down. On R-Car S4 that is a hazard: DBI accesses
> > issued within a few hundred microseconds of an unexpected link down do
> > not complete and hang the host. In testing, the first Root Port config
> > read after powering off the link partner hung unless delayed by ~300 us.
> 
> Out of curiosity, do they trigger SError, and does the firmware (TFA)
> trap/fix those up in EL3?

I haven't confirmed whether those DBI accesses trigger an SError.
The BL31 on the Spider I'm using should be based on rcar-s4_v2.5 [1] from the
Spider BSP (as I haven't updated it). I just checked that this version also sets
HANDLE_EA_EL3_FIRST=1, plus plat_ea_handler() is empty. So IIUC an SError would
be taken to EL3, and the normal path would return to the kernel without fixing
the underlying error.

And, regardless of whether it's a synchronous abort or an SError:
- There was no crash report (via report_unhandled_exception) on the console.
- Once that DBI access in the small window hangs, it never recovers,
  whereas adding udelay(300) before the access avoids the hang. So I
  doubt it's just retrying the same access after a transient synchronous
  abort as well (though I can't rule it out).

[1] https://github.com/renesas-rcar/arm-trusted-firmware/tree/rcar-s4_v2.5

> 
> > Use the pre-MSI callback to check the APP reset status before DBI is
> > touched. When a reset request is latched, mask the sources, ack the
> > request and schedule recovery work. The work calls
> > pci_host_handle_link_down(), which runs the AER-style recovery and
> > resets the controller through reset_root_port(). If the reset fails, the
> > sources stay masked so nothing touches the unrecovered controller.
> > 
> > Only unmasked status bits are handled and pending latches are cleared
> > when re-arming, so requests recorded during probe or the reset itself do
> > not trigger another recovery. Teardown only disables link-down
> > detection: MSI delivery has to keep working while devices are removed.
> > 
> > When iMSI-RX is not used (external MSI controller or pci=nomsi), the
> > DesignWare core does not request intreq_pcim_sub, so request it in the
> > driver.
> 
> Would it make sense to request the line unconditionally, to simplify the
> driver(s) ?

Yes. The pcie-rcar-gen4 can set pp->msi_irq[0] = -ENODEV just like spear13xx,
dra7xx and keembay for the same reason. That also removes the need for adding
.pre_msi_irq addition in patch 4. Some other adjustments might be needed, but I
believe that is the right and clean direction. I will respin with that in mind.

Thanks for the review, that helps a lot!

Best regards,
Koichiro Den

> 
> I also have to wonder, is this specific to R-Car or could it be this is a
> generic property of the DWC controller and this should go into the DWC core
> ?
> 
> Thank you for your help !